An ideal solution to the treatment or cure of type 1 diabetes mellitus would be the formation of new functioning ?-cells from the patient?s own tissues that are not attacked by the autoimmunity, thereby avoiding the need for any immunosuppression. Abundant recent data have suggested that ?-cells are a viable potential source for endogenous transdifferentiation into ?-cells. Here, we describe a pancreatic intraductal viral delivery system in the mouse, wherein a single infusion of an adeno-associated virus (AAV), carrying a pdx1/mafA expression vector, is given to a toxin-induced (alloxan) diabetic mouse. This AAV gene therapy induced robust and durable ?-cell transdifferentiation into ?-cell-like cells through neogenesis, with recovery of over 60% of the ?-cell mass within 4 weeks, and with persistent, durable euglycemia. Serendipitously, when this ?-cell-like cell neogenesis was similarly induced in the autoimmune NOD mouse model, the mice became euglycemic for 4 months or more, without any additional therapy or immunosuppression. To our knowledge, no clinically applicable ?-cell replacement therapy in NOD mice has been successful without immunosuppression. We suspect that the neogenic ?-cell-like cells may not be attacked by the autoimmunity because they are ?imperfect? ?-cells by RNA-seq analysis. Since pancreatic duct injection is routinely performed in humans as a relatively simple, non-surgical procedure, and since numerous viral gene therapy trials are currently ongoing for several diseases, we feel that our approach may be rapidly translatable to humans with type 1 diabetes mellitus. In this proposal, we will perform important proof-of-principle studies in non-human primates (NHP?s) as last steps in preparation for human gene therapy clinical trials. The primate pancreas has a very different texture and consistency than the mouse pancreas (and is very similar to the human pancreas). Thus, the mechanics of the viral delivery will likely require substantial alterations. We had initial problems in the NHP?s with severe hypoglycemia. We have now solved that problem with some preliminary success in a diabetic NHP model. Further studies will investigate this pancreatic ductal infusion approach in the context of AAV neutralizing antibodies. We will also perform detailed analyses of the new ?-cell-like cells, including physiology, gene expression phenotype, and anatomy. In summary, we feel that the proposed studies, if successful, should position us well in preparation for clinical trials in humans with type 1 diabetes mellitus.

Public Health Relevance

Although great efforts have been made in the field of beta cell replacement therapy, the glaring concern for Type I diabetes is that even if we do develop a way to perform such replacement therapy, those replacement cells would seemingly be doomed to the same autoimmune attack as the original beta cells. Here, we describe a possible solution to this problem. We used a viral intraductal infusion strategy to induce alpha cells in the islets of either alloxan-ablated or NOD (autoimmune) diabetic mice to transdifferentiate into beta cells. This transdifferentiation was robust, leading to recovery of 60% of the beta cell mass in the alloxan-treated mice and a 190-fold increase in beta cell mass in NOD mice. This recovery in the NOD mice lasted 4 months, despite no other interventions, including no immunosuppression. We have now developed a system for similar infusions in non-human primates. Here, in this proposal, in an effort to move very close to human gene therapy clinical trials, we will perform extensive analyses in non-human primates. We will strive to optimize the viral construct and delivery method, and then analyze the biology of these neogenic beta cell-like cells and their milieu.

National Institute of Health (NIH)
National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)
Research Project (R01)
Project #
Application #
Study Section
Cellular Aspects of Diabetes and Obesity Study Section (CADO)
Program Officer
Arreaza-Rubin, Guillermo
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Pittsburgh
Schools of Medicine
United States
Zip Code